Radiation Oncology/Palliation/Brain Metastases/Overview
Brain Metastases Overview
- Estimated 170,000 - 200,000 new cases in US per year
- Incidence increasing, likely due to improvements in systemic therapy and greater use of MRI
- On autopsy, ~10-30% of cancer patients have brain metastases
- Lung cancer 34%, breast cancer 30%, melanoma 72% 
- Metastatic brain tumors 10x more common than primary brain tumors
- Single metastasis = only one lesion in the brain, regardless of extra-cranial status
- Solitary metastasis = CNS metastasis as the only site of disease
- No official staging system exists
Graded Prognostic Assessment (GPA)
Utilize www.brainmetGPA.com for most updated information. Since Sperduto 2012, there have been at least 5 Sperduto papers with updates which are all accounted for in the GPA Index calculator. Looks like last update was in 2017. Bookmark using Safari on your phone to add to homepage as an app.
- All sites now stratify for age (except RCC).
- All sites now stratify for presence of extracranial disease.
- All sites now stratify for number of brain mets:
- Lung cancer and RCC now 1-4 / 5+ BM.
- Melanoma and GI now 1 / 2-3 / 4+ BM.
- Breast now 1 / 2+ BM.
Lung cancer now includes histology (AC including EGFR/ALK vs. SqCC), # BM now (1-4 / 5+).
- MS up to 4y if EGFR/ALK+, 1-4 BM only, any age KPS 90+. Note: These numbers likely do not reflect Osimertinib as first line.
- MS up to 2y if non-targetable AC, 1-4 BM only, any age KPS 90+
- MS up to 1y if SqCC, 1-4 BM only, any age KPS 90+.
RCC now includes Hgb ( > 12 g/dL), # of BM now (1-4 / 5+). Note: The only DS-GPA not accounting for age.
- MS up to 3y if 1-4 BM only, Hgb ≥ 11.5, KPS 90+.
Melanoma now includes BRAF positive, # of BM (1 / 2-3 / 4+) and presence of extracranial dz.
- MS up to 3y if solitary brain met and BRAF positive.
Breast cancer now includes number of brain mets (1 / 2+) and presence of extracranial dz.
- MS up to 3y for HER2+/LumB/TP if solitary brain met, any age KPS 90+.
- MS up to 2y for LumA if solitary brain met, any age KPS 90+.
- MS up to 2y for TN if solitary brain met, < 60y and KPS 70+.
- MS up to 1y for TN if solitary brain met and any age KPS 70+.
GI now includes number of brain mets (1 / 2-3 / 4+) and presence of extracranial dz.
- MS up to 1.5y for solitary brain met and KPS 90+.
- 2012 (1993-2010) PMID 22203767 -- "Summary report on the graded prognostic assessment: an accurate and facile diagnosis-specific tool to estimate survival for patients with brain metastases." (Sperduto PW, J Clin Oncol. 2012 Feb 1;30(4):419-25. )
- Retrospective. 3,940 pts from 11 institutions. Updated GPA.
- 2009 (1985-2007) PMID 19942357 -- "Diagnosis-Specific Prognostic Factors, Indexes, and Treatment Outcomes for Patients with Newly Diagnosed Brain Metastases: A Multi-Institutional Analysis of 4,259 Patients." (Sperduto PW, Int J Radiat Oncol Biol Phys. 2009 Nov 24. [Epub ahead of print])
- Retrospective. 4,259 patients from 11 institutions. Development of Diagnosis-Specific GPA (DS-GPA)
- Outcome: For NSCLC/SCLC need 4 prognostic factors (age, KPS, ECM, number of brain mets). For renal cell/melanoma need 2 prognostic factors (KPS/number of brain mets). For breast/GI need 1 prognostic factor (KPS)
- Conclusion: Significant prognostic factors varied by diagnosis.
- 2008 PMID 17931798 -- "A new prognostic index and comparison to three other indices for patients with brain metastases: an analysis of 1,960 patients in the RTOG database." (Sperduto PW, Int J Radiat Oncol Biol Phys. 2008 Feb 1;70(2):510-4. Epub 2007 Oct 10.)
- Retrospective. 1960 patients enorolled on 5 RTOG trials (7916, 8528, 8905, 9104, 9508). GPA index developed. Comparison of existing indices (GPA vs RPA vs Basic Score vs SIR)
- Outcome: Median OS by GPA 0-1 points 2.6 months; GPA 1.5-2.5 points 3.8 months; GPA 3 points 6.9 months; GPA 3.5-4 points 11.0 months (SS). RPA and GPA best discrimination
- Conclusion: GPA as prognostic as RPA and least subjective/easiest to use
Recursive Partitioning Analysis (RPA)
- 1997 RTOG (1979-1993) - PMID 9128946 — "Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials." Gaspar L et al. Int J Radiat Oncol Biol Phys. 1997 Mar 1;37(4):745-51.
- Analysis of factors related to outcome in previous RTOG brain metastasis trials (RTOG 79-16, 85-28, 89-05). About 1/2 of pts had solitary mets.
|Stage||Characteristics||Median Survival (mo)|
|I||KPS >=70, age <65, primary controlled, no other extracranial mets||7.1|
- RPA validation study using RTOG 91-04 - PMID 10863071
- RPA does not predict survival with 4 or more brain metastases (Germany) - PMID 12540980
Number of Mets
- RTOG 7916 PMID 8420868 -- CT characteristics of patients with brain metastases treated in RTOG study 79-16. (Swift PS, Int J Radiat Oncol Biol Phys. 1993 Jan 15;25(2):209-14.)
- Pre- and post-treatment CT analysis for prognostic factors
- Improved survival: <=3 mets (especially for RPA Stage I patients)
Natural history (no treatment)
Duration of survival with symptomatic mets:
- Corticosteroids should be used at onset for all symptomatic patients
- If stable, can start with 8mg BID dexamethasone (or equivalent)
- If whole brain RT, should receive steroids at least 48 hours before treatment to decrease chance of herniation (works well for weekend call)
- Two dosing schedules to consider
- start 8mg BID, begin taper during Week 2 of RT by 2-4mg every fifth day. If symptoms (tumor or steroid withdrawal), go back up a level for 4-8 days, then restart taper. If symptoms after steroids discontinued, restart full regimen (Posner JB, Neurologic Complications of Cancer, Philadelphia: FA Davis, 1995:37, 77, 311)
- start 8mg BID x4 days, then 4mg BID x4 days, then 2mg BID until treatment completed (PMID 1726656)
- AANS/CNS; 2010 PMID 19957014 -- "The role of steroids in the management of brain metastases: a systematic review and evidence-based clinical practice guideline." (Ryken TC, J Neurooncol. 2010 Jan;96(1):103-14. Epub 2009 Dec 3.)
- Asymptomatic patients: insufficient evidence
- Mild symptoms: dexamethasone 4-8 mg/day
- Moderate/severe symptoms: dexamethasone 16 mg/day
- Duration: Slow taper over 2 week period or longer
- Prophylaxis against seizures is not recommended.
- A Meta-Analysis showed lack of efficacy for seizure prophylaxis
- Mayo - PMID 15595331, 2004 — "Seizure prophylaxis in patients with brain tumors: a meta-analysis." Mayo Clin Proc. 2004 Dec;79(12):1489-94.
- 2010 AANS/CNS Guidelines do not recommend routine prophylactic use of anticonvulsants (PMID 19957015)
When to Treat?
- Consider no RT if expected survival < 30 days and KPS <=60
- Expected median survival <3 months
- prostate: hormone-resistant
- breast: failure of 3rd line chemotherapy
- lung: failure of 1st line chemotherapy
- colon: ascites or liver failure
- liver mets and doubling LFTs q month
- SVC syndrome
- HIV with CD4 <25 and RNA >100K
- CHF Class IV with EF <20%
- Creatinine >8
- pO2 <55 and pCO2 >50
- Wisconsin, 2007 PMID 17401015 -- "Regression after whole-brain radiation therapy for brain metastases correlates with survival and improved neurocognitive function." (Li J, J Clin Oncol. 2007 Apr 1;25(10):1260-6.)
- Subset analysis of PIII trial. 135 patients had evaluation of neurocognitive function. "Good responders" defined as tumor shrinkage >45%, "poor responders" tumor shrinkage <45% as a function of population mean
- Conclusion: Tumor shrinkage correlates with survival and neurocognitive preservation. Neurocognitive function stable/improved in long-term survivors. Tumor progression affects neurocognitive function more than WBRT
- Whole brain RT, given in 30 Gy in 10 fractions, remains the generic standard treatment. However, for those with expected longer survival (>6 months), neurocognitive sequelae can be severe with schedules >=3 Gy/fx
- For patients with a single met, especially if their extra-cranial disease is controlled, local treatment (SRS or surgery) and WBRT provides survival benefit, as well as local control benefit
- The role of SRS alone has not yet been established
- For patients with multiple mets, the optimal management has not yet been established
- Algorithm for initial treatment of brain mets from PMID 17673619 (See Review below)
- If KPS <70: WBRT
- If KPS >=70
- If active extracranial disease
- 1 met: WBRT +/- SRS
- 2+ mets: WBRT
- If controlled extracranial disease
- 1 met: Surgery +/- WBRT or SRS +/- WBRT
- 2-3 mets: WBRT +/- Surgery or WBRT +/- SRS
- >3 mets: WBRT
- If active extracranial disease
Patterns of relapse
- PMID 9614957, 1998 — "Patterns of relapse and late toxicity after resection and whole-brain radiotherapy for solitary brain metastases." Nieder C et al. Strahlenther Onkol. 1998 May;174(5):275-8.
- Retrospective. 66 pts treated with surgery -> WBRT. Relapse 27% after 1 yr, 55% after 2 yrs. High local relapse rate for melanoma, non-breast adenoca, and squamous cell. Failure elsewhere in the brain was influnced by systemic disease activity. Size did not correlate with local failure.
Usually with radiosurgery, but few pts may require reirradiation of the whole brain or partial brain. 8 Gy in 2 weeks to 30.6 Gy in 3 weeks.
Whole Brain Reirradiation:
- 1980: Rush Presbyterian (Chicago) PMID 7448697 -- "Results of re-irradiation for cerebral metastases." (Kurup P, Cancer. 1980 Dec 15;46(12):2587-9.)
- 56 pts. Retreatment dose 20 Gy / 10 fractions. Some pts received multiple re-treatments.
- Good response in 18 pts(32%), partial response in 24(42%). Median duration of response 2.5 months. Median survival 3.5 months.
- 1988: Colorado PMID 2841266 -- "Brain metastases: results and effects of re-irradiation." (Hazuka MB, Int J Radiat Oncol Biol Phys. 1988 Aug;15(2):433-7.)
- 44 pts. Varying doses. Retreatment: median 25 Gy / 3.0 Gy per fraction (range: 6-36 Gy, 2-4 Gy/fx). 1st course: 30/10 (median).
- Median survival 8 weeks. 27% showed partial neurological improvement; 55% failed to respond or had worsening function. Brain necrosis in 3 pts at autopsy.
- Conclusion: "It is concluded that retreatment of brain metastases is seldom worthwhile. Survival is usually short and most importantly, the quality of survival frequently is not improved."
- 1990: NYU PMID 2305074 -- "Cerebral metastases: value of reirradiation in selected patients." (Cooper JS, Radiology. 1990 Mar;174(3 Pt 1):883-5.)
- 52 pts. Retreatment: 25 Gy in 10 fractions. 1st course: 30/10.
- Neuro improvement in 42%. Survival 5 months.
- 1996: Mayo PMID 8621282 -- "Analysis of outcome in patients reirradiated for brain metastases." (Wong WW, Int J Radiat Oncol Biol Phys. 1996 Feb 1;34(3):585-90.)
- 86 pts. Varying doses. Retreatment: median 20 Gy / 10 fractions. 1st course: 30/10 (median).
- Neuro symptoms cleared in 27%, partially improved in 43%, 29% no change or worsened. Median survival 4 months.
- 2011: MGH (2002-2008) PMID 21620583 -- "Outcomes After Whole Brain Reirradiation in Patients with Brain Metastases." (Son CH, Int J Radiat Oncol Biol Phys. 2011 May 26. [Epub ahead of print])
- 17 pts w/ brain mets treated initially with WBRT (median 35 Gy, 2.5 Gy/fx) and given whole brain reirradiation (median 21.6 Gy [range 14-30 Gy]; median 1.8 Gy/fx [range 1.5-2.0 Gy/fx]).
- Of 10 pts w/ complete f/u information, 8 experienced complete or partial improvement in symptoms; 2 did not improve. Time to radiographic progression 5.2 months. Median OS (after initial diagnosis of brain mets) 24.7 mo. Median survival after reirradiation 5.2 mo. In 6 pts with stable extracranial disease, MS after re-RT 19.8 mo vs 2.5 mo with extracranial disease progression. Side effects mild to moderate
- Conclusion: reirradiation of brain mets may be appropriate in selected pts, especially those with controlled extracranial disease.
Reirradiation after PCI: Reirradiation for subsequent brain metastases after prophylactic cranial irradiation
- Institut Gustave-Roussy (1985-93) - trials PCI85 and PCI88
- PMID 12075744, 2002 — "Patterns of failure after prophylactic cranial irradiation in small-cell lung cancer: analysis of 505 randomized patients." (Arriagada R, Ann Oncol. 2002 May;13(5):748-54.)
- Analysis of two randomized trials (+/- PCI) looking at patterns of failure
- Overall 5-yr rates of brain metastases were 59% (PCI-) and 43% (PCI+). PCI dose was 24-30 Gy (with <= 3Gy/fx); commonly 24 Gy (3Gy x 8) was used.
- Per protocol, patients developing brain metastases were treated with 50 Gy in 28 fractions if no prior PCI or 39 Gy in 22 fractions (approx 1.77 Gy/fx) if prior PCI.
- Harvard; 2007 PMID 17673619 -- "Multidisciplinary management of brain metastases." (Eichler AF, Oncologist. 2007 Jul;12(7):884-98.)
- Wisconsin; 2006 PMID 16525185 -- "Whole-brain radiotherapy in the management of brain metastasis." (Khuntia D, J Clin Oncol. 2006 Mar 10;24(8):1295-304.)
- Please see the Brain Treatment Toxicity section
- Please see the Acute Parotitis section
- Please see the Phenytoin-Induced Dermatitis section
- Wake Forest, 2007 PMID 17869448 -- "A Phase III, Double-Blind, Placebo-Controlled Prospective Randomized Clinical Trial of d-threo-methylphenidate HCl in Brain Tumor Patients Receiving Radiation Therapy." (Butler JM Jr, Int J Radiat Oncol Biol Phys. 2007 Sep 13; [Epub ahead of print])
- Randomized. 68 patients with primary or metastatic brain tumors. Arm 1) prophylactic d-MPH 5mg BID -> 15mg BID vs. Arm 2) observation. Outcome measure QOL and cognitive function
- Outcome: No difference in fatigue, QOL, or cognitive function
- Conclusion: Prophylactic d-MPH not beneficial in brain tumor patients
- 1989 MSKCC PMID 2725874 -- Radiation-induced dementia in patients cured of brain metastases. (DeAngelis LM, Neurology. 1989 Jun;39(6):789-96.)
- Retrospective. 12 patients with delayed complications of WBRT
- WBRT given: total dose only 25-39 Gy, but given in 3-6 Gy/fx
- Complications (median 14 months, 5-36 months): progressive dementia, ataxia, urinary incontinence causing severe disability; death in 7/12 patients (from side effects, not tumor recurrence). CT scan: cortical atrophy, radionecrotic lesions
- Conclusion: incidence of severe WBRT-induced dementia 2-5%. Fraction too high.
- RTOG 0614, 2013 PMID 23956241 -- "Memantine for the prevention of cognitive dysfunction in patients receiving whole-brain radiotherapy: a randomized, double-blind, placebo-controlled trial." (Brown PD, Neuro Oncol 2013 Aug 16; [Epub ahead of print])
- Randomized. 554 patients treated for CNS metastatic disease with whole brain RT (3750 cGy at 250 cGy/fx) randomized to 1) Memantine during and after RT (total 24 weeks) or 2) placebo.
- Primary Outcome: Less decline in delayed recall with memantine but borderline statistical significance (p=0.059). Probably related to loss of study power over time.
- Secondary Outcomes: Time to cognitive failure improved with memantine (54% vs. 65% at 24 weeks; p=0.01). No difference in PFS or OS.
- Toxicity: No difference. 28% grade 3-4, half of which was attributable to therapy (same in each arm).
- Conclusion: Patients treated with memantine had better cognitive function over time; specifically, memantine delayed time to cognitive decline and reduced the rate of decline in memory, executive function, and processing speed in patients receiving WBRT. Primary endpoint not technically met, perhaps due to patient loss over time.
- Radiation Oncology:Melanoma - data related specifically to brain mets from melanoma
- Radiation Oncology:Renal Cell - data related specifically to brain mets from renal cell CA
- Radiation Oncology:Thromboembolism - information on anticoagulation and brain mets
- Oncology - Management of Brain Metastases. Wen PJ and Loeffler JS. Vol 13, No 7 (July 1999)
- ^ - PMID 13172684 - "Roentgen-ray therapy of cerebral metastases." Cancer. 1954 Jul;7(4):682-9.
- ^ - PMID 718475 - "Treatment for patients with cerebral metastases." Markesbery WR et al. Arch Neurol. 1978 Nov;35(11):754-6.
- ^ - PMID 14264031 - "USE OF GLUCOCORTICOIDS IN THE PALLIATIVE TREATMENT OF METASTATIC BRAIN TUMORS." Ruderman NB et al. Cancer. 1965 Mar;18:298-306. (No abstract online)
- ^ - PMID 570349 (No abstract) - "Intracranial metastases from systemic cancer." Posner JB et al. Adv Neurol. 1978;19:579-92.